Investigation of Monte Carlo uncertainty prediction of CO concentration via artificial neural network

سال انتشار: 1401
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 217

فایل این مقاله در 15 صفحه با فرمت PDF قابل دریافت می باشد

این مقاله در بخشهای موضوعی زیر دسته بندی شده است:

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

ICRSIE07_155

تاریخ نمایه سازی: 6 اردیبهشت 1402

چکیده مقاله:

In this study, daily meteorological data (average daily temperature, precipitation, pressure, humidity, wind speed, horizontal visibility, and traffic data) were used to determining the uncertainty of the Monte Carlo CO gas concentration prediction in Tehran. Accordingly, an artificial neural network was used for prediction and the information recorded in ۲۰۱۶ was considered as input to the network. To do this, Backpropagation (BP) algorithm was used to achieve the best neural network performance using a multilayer perceptron (MLP) method. The output results were compared with each other using the root mean square error (RSME) and it was found that the three-layer neural network with ۷ neurons in the first layer and ۵ neurons in the second had the best performance. The RSME value, in this case, was equal to ۱.۲۶ and using the Monte Carlo simulation method, the uncertainty was estimated at ۹۵% confidence level. Also, sensitivity analysis and correlation coefficients of the parameters have been obtained as a criterion for measuring their effects. The results of this study showed that, the concentration of CO pollutants was estimated to be between ۴۳.۱۲۵ and ۴۳.۷۰۳, also, the temperature parameter has the greatest impact on the concentration of this pollutant in Tehran. Therefore, by using the air pollution prediction model in Tehran metropolis and uncertainty and sensitivity analysis, the trend of CO change can be informed in advance and with management strategies within the framework of preventive measures such as traffic restrictions at predetermined times, prevent excessive air pollution, which significantly reduces financial and human damages.

نویسندگان

Mohsen Taghavijeloudar

Department of Civil and Environmental Engineering, Seoul National University, Seoul, Korea

Saba Abdolalian

MSc of Civil and Environmental Engineering, Faculty of Civil Engineering, Babol NoshirvaniUniversity of Technology, Babol, Iran,

Babak keshvari

BSc of Civil Engineering, Faculty of Civil Engineering, university of Mazandaran, Babolsar, Iran